Resistor Network Research Articles

Form-finding using the Force Density Method: Existence of solutions, singularities, and an analogy to electric circuits

The Force Density Method is the oldest computational method for form-finding of funicular structures. The method is based on solving a linear system of equilibrium equations. Linearity ensures existence and uniqueness of the solution, provided that the coefficient matrix is non-singular. In the opposite case, the method often fails to deliver a feasible solution. In this study, we highlight a mathematical analogy between force density networks and electric circuits. Motivated by standard reduction steps in networks of electric resistors, inductors, or capacitors, we develop a new sequential reduction technique applicable to networks of force density elements. Network reduction enables analytic detection of singularity in many cases. Several examples illustrate the feasibility of the proposed singularity test, furthermore the role of singularities in classification of emergent structural forms is discussed.

Two methods for thermal conduction analysis of bi-porous sintered structures

Loop heat pipes (LHPs) have consistently garnered attention for effectively managing thermal conditions in critical electronic components of spacecraft and satellites. Particularly intriguing are LHPs equipped with bi-porous capillary wicks, demonstrating commendable flow and thermal dissipation capabilities. Despite its significance for evaluating the phase-transition heat transfer in porous structures, the effective thermal conductivity (ETC) has not been understood mechanistically. This study aims to establish an efficient procedure for predicting ETC in bi-porous wicks through a comparative analysis of analytical and numerical methods. The analytical ETC expression was formulated using thermal resistance network and neck formation theory. Meanwhile, the numerical model utilized the Finite volume method (FVM) combined with a 3D porous structure created by Diffusion-Limited Aggregation (DLA) algorithm. The investigation revealed a congruence between two methods and the measured ETCs of interstitial-pore samples containing fine nickel powders. However, undesirable results were shown for coarse samples due to the irregular particle shapes. In samples with formation pores, the numerical model excels in characterizing the spatial distribution and scale of pores, albeit at the expense of time consumption and model complexity compared to the analytical one. Both methods prove valuable for ETC modeling of sintered bi-porous structures according to desired applications.

Ultra-Fast Ion Mobility Spectrometer for High-Throughput Chromatography.

Fast chromatography systems especially developed for high sample throughput applications require sensitive detectors with a high repetition rate. These high throughput techniques, including various chip-based microfluidic designs, often benefit from detectors providing subsequent separation in another dimension, such as mass spectrometry or ion mobility spectrometry (IMS), giving additional information about the analytes or monitoring reaction kinetics. However, subsequent separation is required at a high repetition rate. Here, we therefore present an ultra-fast drift tube IMS operating at ambient pressure. Short drift times while maintaining high resolving power are reached by several key instrumental design features: short length of the drift tube, resistor network of the drift tube, tristate ion shutter, and improved data acquisition electronics. With these design improvements, even slow ions with a reduced mobility of just 0.94 cm2/(V s) have a drift time below 1.6 ms. Such short drift times allow for a significantly increased repetition rate of 600 Hz compared with previously reported values. To further reduce drift times and thus increase the repetition rate, helium can be used as the drift gas, which allows repetition rates of up to 2 kHz. Finally, these significant improvements enable IMS to be used as a detector following ultra-fast separation including chip-based chromatographic systems or droplet microfluidic applications requiring high repetition rates.

Engineering the structural and electrical interplay of nanostructured Au resistive switching networks by controlling the forming process

Networks of random-assembled gold clusters produced in the gas phase show resistive switching (RS) activity at room temperature and they are suitable for the fabrication of devices for neuromorphic data processing and classification. Fully connected cluster-assembled nanostructured Au films are characterized by a granular structure rich of interfaces, grain boundaries and crystalline defects. Here we report a systematic characterization of the electroforming process of the cluster-assembled films demonstrating how this process affects the interplay between the nano- and mesoscale film structure and the neuromorphic characteristics of the resistive switching activity. The understanding and the control of the influence of the resistive switching forming process on the organization of specific structures at different scales of the cluster-assembled films, provide the possibility to engineer random-assembled neuromorphic architectures for data processing task.

CNSC-32. SIMULTANEOUS INVESTIGATION OF TUMOR CELL BEHAVIOR AND CONNECTIVITY WITH CORRELATIVE IN VIVO TWO-PHOTON AND EX VIVO EXPANSION MICROSCOPY

Abstract Glioblastomas are the most aggressive primary brain tumors, characterized by whole-brain colonization, high therapeutic resistance, and inevitable recurrence. It has previously been shown that glioblastoma cells can form functionally gap-junction connected tumor cell networks with other tumor cells and/or astrocytes, which play a crucial role in tumor progression and treatment response. Furthermore, there is increasing evidence that synaptic neuron-glioma networks promote tumor growth. However, methodologies to investigate the direct link between synaptic, astrocytic and tumor cell connectivity together with tumor cell behavior including invasion, proliferation and therapeutic resistance are yet lacking. We aimed to characterize neuron-tumor and tumor-tumor/astrocyte cell interactions on a single cell level by using correlative in vivo two-photon and ex vivo expansion microscopy with near synaptic-resolution. We studied glioblastoma cells with intravital two-photon microscopy using patient-derived glioblastoma mouse models and the fluorescent functional imaging dye Sulforhodamine 101 allowing to investigate intratumoral and astrocytic connectivity over an extended period up to weeks. Subsequently, these functional tumor cell subpopulations could be directly correlated ex vivo to identify neuron-glioma synaptic connectivity. We utilized machine learning based segmentation and a semi-automated analysis pipeline to explore the connection of different glioblastoma morphological and behavioral subtypes to direct and indirect neuronal input in multiple xenograft models. With this new correlative method, we were able to simultaneously evaluate tumor cell behavior, synaptic and tumor-tumor/astrocyte connections on a single cell level. In addition, we demonstrated that this pipeline can be used for patient samples, revealing its clinical-translational potential for cancer neuroscience. In summary, we have established a new correlative technology that interlinks two key aspects of brain cancer neuroscience: cancer cell behavior and brain tumor network connectivity. This methodology promises a more thorough understanding of glioblastoma biology, facilitating the discovery of cancer biology, including the role of brain tumor networks in therapeutic resistance.

Development of robust and accurate thermo-electrochemical models for Lithium-ion batteries

Integrating lithium-ion batteries (LIBs) as an energy storage system in electronic devices, including electric vehicles (EVs), creates new challenges. LIB design must be optimized according to the specifications of each application to improve battery performance and safety in each application while avoiding rapid battery degradation. Therefore, accurate models that can forecast the performance of LIBs under various operating conditions are required. Consequently, the primary purpose of this study is to establish a battery model that accounts for the correct thermal heating of large prismatic batteries (PBs) used in EVs. Due to large fluctuations in the space current and Temp (Temp) of PCs (PCs), one-dimensional (1D) models cannot be used to model these PCs. This work combines a 1D model into a three-dimensional multilayer model by the resistive network method. The established model was validated by evaluating the simulated surface Temp of a PB (containing LFP as the positive (+ve) electrode and graphite as the negative (-ve) electrode) with experimental data at different charge (CH)/ discharge (DISCH) rates (1C, 2C, 3C, and 5C). Application of the model utilized effect size, location, and thickness of current collector sheets on the electrochemical (electrochem) properties of large cells is evaluated. Moving the bumps from the edge and one side (a standard commercial design) to the center and the other side of the cell and maximizing their width have been shown to reduce the change in imbalance caused by electrochem current flow.

Ribosome profiling reveals the fine-tuned response of Escherichia coli to mild and severe acid stress.

Bacteria react very differently to survive in acidic environments, such as the human gastrointestinal tract. Escherichia coli is one of the extremely acid-resistant bacteria and has a variety of acid-defense mechanisms. Here, we provide the first genome-wide overview of the adaptations of E. coli K-12 to mild and severe acid stress at both the transcriptional and translational levels. Using ribosome profiling and RNA sequencing, we uncover novel adaptations to different degrees of acidity, including previously hidden stress-induced small proteins and novel key transcription factors for acid defense, and report mRNAs with pH-dependent differential translation efficiency. In addition, we distinguish between acid-specific adaptations and general stress response mechanisms using denoising autoencoders. This workflow represents a powerful approach that takes advantage of next-generation sequencing techniques and machine learning to systematically analyze bacterial stress responses.

Mesoscopic Conductance Fluctuations in 2D HgTe Semimetal.

Mesoscopic conductance fluctuations were discovered in a weak localization regime of a strongly disordered two-dimensional HgTe-based semimetal. These fluctuations exist in macroscopic samples with characteristic sizes of 100 μm and exhibit anomalous dependences on the gate voltage, magnetic field, and temperature. They are absent in the regime of electron metal (at positive gate voltages) and strongly depend on the level of disorder in the system. All the experimental facts lead us to the conclusion that the origin of the fluctuations is a special collective state in which the current is conducted through the percolation network of electron resistances. We suppose that the network is formed by fluctuation potential whose amplitude is higher than the Fermi level of electrons due to their very low density.

Green's functions for random resistor networks.

We analyze random resistor networks through a study of lattice Green's functions in arbitrary dimensions. We develop a systematic disorder perturbation expansion to describe the weak disorder regime of such a system. We use this formulation to compute ensemble-averaged nodal voltages and bond currents in a hierarchical fashion. We verify the validity of this expansion with direct numerical simulations of a square lattice with resistances at each bond exponentially distributed. Additionally, we construct a formalism to recursively obtain the exact Green's functions for finitely many disordered bonds. We provide explicit expressions for lattices with up to four disordered bonds, which can be used to predict nodal voltage distributions for arbitrarily large disorder strengths. Finally, we introduce a novel order parameter that measures the overlap between the bond current and the optimal path (the path of least resistance), for a given resistance configuration, which helps to characterize the weak and strong disorder regimes of the system.

Experimental study on the thermal performance of a liquid-cooled heat sink integrating heat pipes for dual CPU servers

The increasing power consumption of server CPUs poses a challenge for data center cooling. In this work, a liquid-cooled heat pipe heat sink (LHPHS) with compactly arranged heat pipes is designed and fabricated for CPU cooling in dual CPU servers. Water is selected as the working fluid in the heat pipes. Experiments are conducted to explore the heat dissipation potential of LHPHS under high inlet coolant temperature and high heat load conditions, and to investigate the operational stability of LHPHS under the conditions of uneven heat generation between two CPUs. Furthermore, a one-dimensional thermal resistance network and numerical model of the LHPHS are established to quantitatively analyze the thermal resistance distribution of LHPHS and the variation of thermal resistance with coolant flow rate. The results show that, firstly, the LHPHS operates stably under high inlet coolant temperature and high heat load conditions, meeting the heat dissipation requirements of up to 37.5 W/cm2 with a coolant flow rate of 1 L/min and inlet coolant temperature of 35℃. Secondly, the LHPHS exhibits excellent performance in controlling temperature difference between CPUs and even when there is discrepant heat generation between them. Thirdly, the LHPHS performs a low thermal resistance, measuring 0.044 ℃/W at a coolant flow rate of 0.5 L/min, and this resistance further decreases as the flow rate increases. Additionally, it is observed that the thermal resistance of the liquid-cooled plate decreases with the increasing flow rate, and heat transfer within the heat pipes becomes a critical aspect under high coolant flow rates. Therefore, reducing the thermal resistance of the heat pipe is crucial for the thermal performance improvement of the LHPHS.

Comparative transcriptome analysis reveals key pathways and regulatory networks in early resistance of Glycine max to soybean mosaic virus.

As a high-value oilseed crop, soybean [Glycine max (L.) Merr.] is limited by various biotic stresses during its growth and development. Soybean mosaic virus (SMV) is a devastating viral infection of soybean that primarily affects young leaves and causes significant production and economic losses; however, the synergistic molecular mechanisms underlying the soybean response to SMV are largely unknown. Therefore, we performed RNA sequencing on SMV-infected resistant and susceptible soybean lines to determine the molecular mechanism of resistance to SMV. When the clean reads were aligned to the G. max reference genome, a total of 36,260 genes were identified as expressed genes and used for further research. Most of the differentially expressed genes (DEGs) associated with resistance were found to be enriched in plant hormone signal transduction and circadian rhythm according to Kyoto Encyclopedia of Genes and Genomes analysis. In addition to salicylic acid and jasmonic acid, which are well known in plant disease resistance, abscisic acid, indole-3-acetic acid, and cytokinin are also involved in the immune response to SMV in soybean. Most of the Ca2+ signaling related DEGs enriched in plant-pathogen interaction negatively influence SMV resistance. Furthermore, the MAPK cascade was involved in either resistant or susceptible responses to SMV, depending on different downstream proteins. The phytochrome interacting factor-cryptochrome-R protein module and the MEKK3/MKK9/MPK7-WRKY33-CML/CDPK module were found to play essential roles in soybean response to SMV based on protein-protein interaction prediction. Our findings provide general insights into the molecular regulatory networks associated with soybean response to SMV and have the potential to improve legume resistance to viral infection.

Frequency of isolation and phenotypic antimicrobial resistance of fecal Salmonella enterica recovered from dairy cattle in Canada

Salmonellosis is one of the leading causes of gastrointestinal infections in humans. In Canada, it is estimated that approximately 87,500 cases of salmonellosis occur every year in humans, resulting in 17 deaths. In the United States, it is estimated that 26,500 hospitalizations and 420 deaths occur every year. In dairy cattle, infections caused by nontyphoidal Salmonella enterica can cause mild to severe disease, including enteritis, pneumonia, and septicemia. Our study objectives were to determine the proportion of fecal samples positive for Salmonella in dairy cattle in Canada and determine the resistance pattern of these isolates. We used data collected through the Canadian Dairy Network for Antimicrobial Stewardship and Resistance (CaDNetASR). Pooled fecal samples from preweaning calves, postweaning heifers, lactating cows, and manure storage were cultured for Salmonella, and the isolates were identified using matrix-assisted laser desorption/ionization-time of flight mass spectrometry. Antimicrobial susceptibilities were determined using the minimum inhibitory concentration test, and resistance interpretation was made according to the Clinical and Laboratory Standards Institute. A 2-level, multivariable logistic regression model was built to determine the probability of recovering Salmonella from a sample, accounting for province, year, and sample source. The proportion of farms with at least one positive sample were 12% (17/140), 19% (28/144), and 17% (24/144) for the sampling years 2019, 2020, and 2021, respectively. Out of the 113 Salmonella isolates, 23 different serovars were identified. The occurrence of Salmonella appeared to be clustered by farms and provinces. The most common serovars identified were Infantis (14%) and Typhimurium (14%). Overall, 21% (24/113) of the Salmonella isolates were resistant to at least one antimicrobial. Resistance to tetracycline was commonly observed (17%); however, very limited resistance to category I antimicrobials (categorization according to Health Canada that includes third-generation cephalosporins, fluoroquinolones, polymyxins, and carbapenems) was observed, with one isolate resistant to amoxicillin and clavulanic acid. The proportion of Salmonella isolates resistant to 2 and 3 antimicrobial classes was 3.5% and 8.8%, respectively. Our study provided valuable information on the proportion of fecal samples positive for Salmonella, the serovars identified, and the associated resistance patterns across CaDNetASR herds, at regional and national levels.

Explicit potential function and fast algorithm for computing potentials in [formula omitted] conic surface resistor network

Resistor network research is of great importance, yet many resistor networks and their large-scale fast computations have not received sufficient attention. This paper proposes a new resistor network with idiosyncratic shape, i.e., a α×β conic surface (CS) resistor network that resembles the upper part of a three-dimensional Dirac function. Utilizing the Recursion Transform (RT-V) method of Tan, a recursive matrix equation model is constructed based on Kirchhoff’s law and nodal voltages, which contains the modified tridiagonal Toeplitz matrix. By using the orthogonal matrix transformation, the eigenvalues and eigenvectors of the modified tridiagonal Toeplitz are obtained. The discrete sine transform of the fourth type (DST−IV) is utilized to solve node voltages, while the explicit potential function is represented by the Chebyshev polynomials of the second kind. In addition, explicit potential functions for some special cases are provided, and the potential distribution is illustrated using dynamic three-dimensional graph. To achieve a rapid calculation of the potential, a fast algorithm based on the multiplication of DST-IV with a vector is proposed. In the end, analysis of computational efficiency for the explicit potential function and the fast algorithm are shown.

Three in One: Superoleophilic, Chemically and Mechanically Resistant ZIF-7 and ZIF-11 Percolation Networks for Selective Permeation of Oils and Chlorinated Solvents.

Imbuing superwetting functions to organic-inorganic hybrid networks displaying chemical resistance, self-cleaning ability, and selective permeation of liquids has received increasing attention in recent years. Here we report superhydrophobic ZIF-7 and ZIF-11 on multilayer fluorinated graphene (FG) nanosheets with long-lasting water-repellent features. By exploring the solution processing of these chemically resistant dispersions, superoleophilic FG-ZIF-7 stainless steel mesh (FG-ZIF-7-SSM) and FG-ZIF-11 over cotton cloth (FG-ZIF-11-CC) possessing superior adhesion were fabricated. These permselective oil-liking prototypes were explored toward mesitylene and crude oil pickup from chemically harsh marine conditions such as seawater, acidic water, and alkaline water, with a separation efficiency of 96-94% up to 10 cycles. Furthermore, using an FG-ZIF-11-CC-wrapped glass pipet, upward diffusion of chloroform from sea, acidic, and alkaline water in 45 s was demonstrated with a separation efficacy of 94% up to 20 cycles. In addition to the chemical resistance and reusability, the mechanical stability of FG-ZIF-7-SSM and FG-ZIF-11-CC was investigated through folding, tape peeling, and dragging through sandpaper up to 250 cycles, showing no signs of changes in the hydrophobic responses. This research sheds light on the application of physiochemically resistant percolation coatings based on fluorinated graphene multilayers supporting ZIF-7 and ZIF-11 toward oil/water separation.

Numerical and experimental study on efficient optimization of variable cross-section battery thermal management systems using an improved flow resistance network model

Air-cooled systems are widely used for cooling of battery packs in electric vehicles. Optimization method combined with the flow resistance network (FRN) model is effective for system design. However, the local pressure loss of divergence and convergence ducts in the traditional FRN model is ignored, leading to large error when applying the model in systems with variable cross-section channels. In this study, the local loss coefficients of the contractive and diffusive channels are derived and the improved FRN model is developed for velocity calculation of variable cross-section air-cooled systems. The improved model is verified using computational fluid dynamics method. Based on the improved FRN model, the cross-section width distribution of the divergence duct is calculated from the given flow rate distribution. The shape of the divergence deflector is optimized by adjusting the flow rate distribution and using the improved model. The results indicate that the temperature difference of the battery cells decreases by more than 84% after the optimization, with the maximum temperature decreasing by more than 3.8 K. Furthermore, the optimized system performs better than those in the previous studies. Also, experiments are carried out to verify the effectiveness of the developed optimization method based on the improved FRN model.

HIV-1 drug resistance and genetic transmission network among newly diagnosed people living with HIV/AIDS in Ningbo, China between 2018 and 2021

BackgroundAs the HIV epidemic continues to grow, transmitted drug resistance(TDR) and determining relationship of HIV transmission are major barriers to reduce the risk of HIV transmissions.This study aimed to examine the molecular epidemiology and TDR and evaluated the transmission pattern among newly diagnosed people living with HIV/AIDS(PLWHA) in Ningbo city, which could contribute to the development of targeted precision interventions.MethodsConsecutive cross-sectional surveys were conducted in Ningbo City between January 2018 and December 2021. The HIV-1 pol gene region was amplified and sequenced for drug resistance and genetic transmission network analysis. TDR was determined using the Stanford University HIV Drug Resistance Database. Genetic transmission network was visualized using Cytoscape with the genetic distance threshold of 0.013.ResultsA total of 1006 sequences were sequenced successfully, of which 61 (6.1%) showed evidence of TDR. The most common mutations were K103N (2.3%), E138A/G/Q (1.7%) and V179D/E (1.2%). 12 HIV-1 genotypes were identified, with CRF07_BC being the major genotype (43.3%, 332/767), followed by CRF01_AE (33.7%, 339/1006). 444 (44.1%) pol sequences formed 856 links within 120 transmission clusters in the network. An increasing trend in clustering rate between 2018 and 2021(χ2 = 9.546, P = 0.023) was observed. The odds of older age (≥ 60 years:OR = 2.038, 95%CI = 1.072 ~ 3.872, compared to < 25 years), HIV-1 genotypes (CRF07_BC: OR = 2.147, 95%CI = 1.582 ~ 2.914; CRF55_01B:OR = 2.217, 95%CI = 1.201 ~ 4.091, compared to CRF01_AE) were significantly related to clustering. Compared with CRF01_AE, CRF07_BC were prone to form larger clusters. The largest cluster with CRF07_BC was increased from 15 cases in 2018 to 83 cases in 2021.ConclusionsThis study revealed distribution of HIV-1 genotypes, and genetic transmission network were diverse and complex in Ningbo city. The prevalence of TDR was moderate, and NVP and EFV were high-level NNRTI resistance. Individuals aged ≥ 60 years old were more easily detected in the networks and CRF07_BC were prone to form rapid growth and larger clusters. These date suggested that surveillance and comprehensive intervention should be designed for key rapid growth clusters to reduce the potential risk factors of HIV-1 transmission.

Predicting mechanical and electrical failure of nanowire networks in flexible transparent electrodes

Flexible transparent electrodes employing metal nanowires (NWs) find extensive use in various applications such as optoelectronic devices, solar cells, light-emitting diodes, and transparent heaters. NW networks in flexible transparent electrodes can withstand mechanical deformations and conduct electricity but are susceptible to localized damage caused by mechanical stress and current density concentration. This localized damage ultimately results in electrode failure. Our study aims to track locally induced damage from both mechanical and electrical sources and assess their collective influence on electrode performance until failure occurs. To this end, we create two-dimensional digital samples that represent the NW networks, transform them into beam networks and equivalent resistor networks, and perform finite element simulations of the mechanical and electrical network responses while varying the NW content.Our simulations reveal crack-like patterns in the distribution of damaged elements at network failure that depend on the process inducing the damage. While our results suggest that the impact of electrically induced damage on overall network stability is more significant than that of mechanically induced damage, the latter must not be ignored.

Is the Urban Landscape Connected? Construction and Optimization of Urban Ecological Networks Based on Morphological Spatial Pattern Analysis

Urban green ecological space is an important measure of sustainable urban development. Among them, landscape connectivity is one of the key factors in maintaining landscape function. Ecological networks can effectively improve regional ecological quality and promote urban landscape connectivity. However, previous studies on ecological networks have mainly focused on biodiversity conservation and lack research on landscape connectivity. This study used morphological spatial pattern analysis methods and utilized connectivity indices to identify ecological sources in the Chaoyang and Nanguan districts of China’s Changchun City and selected environmental and anthropogenic factors to construct an integrated resistance surface. The minimum cumulative resistance model and network structure index were used for urban ecological network construction and node optimization. The results show that the potential ecological network comprises 17 ecological sources and 34 potential corridors, primarily located in forests and water bodies in the east and south regions. However, the northwest has poor habitat quality and uneven distribution of ecological corridors, that warrant prioritization in future planning, construction, and protection efforts. By introducing six supplemental sources and 25 additional corridors, the function and overall connectivity of the regional ecological network can be improved. The study confirmed that the selection of appropriate connectivity thresholds can improve the accuracy of ecological sources identification, and that the influence of anthropogenic factors on ecological resistance cannot be ignored. This study will provide a scientific basis for promoting urban construction and ecological balance.

Designing Smart Legging for Posture Monitoring Based on Textile Sensing Networks

Running is a highly popular form of exercise, while incorrect running posture over an extended period can lead to severe knee injuries. Smart textiles have recently demonstrated significant potential for continuous motion monitoring. This study involved the design and development of a smart legging with a resistive textile sensor network to monitor lower body motion. The study consists of three main parts. Firstly, we tested textile sensors in terms of linearity and robustness to determine the basic sensor unit that can monitor the characteristics of running postures. Next, optimal sensor placement was determined through comparison experiments, and a sensor network was proposed. Finally, based on the LSTM model with data gathered from 6 participants, we developed the smart legging system that is capable of identifying three types of improper running postures and normal postures with 99.1% accuracy. The evaluation revealed that the smart legging system had the potential to help users adjust their running postures to prevent knee injury through continuous monitoring and multi-modal feedback.

Effect of adherence to primaquine on the risk of Plasmodium vivax recurrence: a WorldWide Antimalarial Resistance Network systematic review and individual patient data meta-analysis

BackgroundImperfect adherence is a major barrier to effective primaquine radical cure of Plasmodium vivax. This study investigated the effect of reduced adherence on the risk of P. vivax recurrence.MethodsEfficacy studies of patients with uncomplicated P. vivax malaria, including a treatment arm with daily primaquine, published between January 1999 and March 2020 were identified. Individual patient data from eligible studies were pooled using standardized methodology. Adherence to primaquine was inferred from i) the percentage of supervised doses and ii) the total mg/kg dose received compared to the target total mg/kg dose per protocol. The effect of adherence to primaquine on the incidence of P. vivax recurrence between days 7 and 90 was investigated by Cox regression analysis.ResultsOf 82 eligible studies, 32 were available including 6917 patients from 18 countries. For adherence assessed by percentage of supervised primaquine, 2790 patients (40.3%) had poor adherence (≤ 50%) and 4127 (59.7%) had complete adherence. The risk of recurrence by day 90 was 14.0% [95% confidence interval: 12.1–16.1] in patients with poor adherence compared to 5.8% [5.0–6.7] following full adherence; p = 0.014. After controlling for age, sex, baseline parasitaemia, and total primaquine dose per protocol, the rate of the first recurrence was higher following poor adherence compared to patients with full adherence (adjusted hazard ratio (AHR) = 2.3 [1.8–2.9]). When adherence was quantified by total mg/kg dose received among 3706 patients, 347 (9.4%) had poor adherence, 88 (2.4%) had moderate adherence, and 3271 (88.2%) had complete adherence to treatment. The risks of recurrence by day 90 were 8.2% [4.3–15.2] in patients with poor adherence and 4.9% [4.1–5.8] in patients with full adherence; p < 0.001.ConclusionReduced adherence, including less supervision, increases the risk of vivax recurrence.